Chris Wikle

View Calendar
March 22, 2021 11:30 am - 12:30 pm

Adapting Efficient Reservoir Neural Methods for Statistical Long-Lead Forecasting of Ecological Processes

If it is so difficult to forecast the weather a few days from now, how can we forecast the state of weather-sensitive ecological systems one year out? The atmosphere is a chaotic dynamical system, and because of that, skillful weather forecasts are only possible out to about 7 to 10 days.  However, dynamical processes in the ocean operate on a much longer time scales, and many atmospheric processes depend crucially on the ocean as a forcing mechanism.  Some of these processes can be forced remotely, leading to so-called teleconnections.  This coupling between the slowly varying ocean and the faster varying atmosphere and associated processes, allows for the skillful prediction of some general properties of such systems many months in advance.  These long-lead forecasts are important for activities such as agricultural and wildlife management, energy production, and disaster planning, to name a few.  It has consistently been established that statistical models can provide some of the most skillful long-lead forecasts. However, it can be particularly challenging to specify parameterizations for nonlinear dynamical spatio-temporal forecast models that are simultaneously useful scientifically and efficient computationally, and that allow for proper uncertainty quantification.  In some cases, when such information is available, one can embed mechanistic information into multi-level (deep/hierarchical) models to facilitate parameter reduction and interpretability.  When such information is not available (such as with animal behavior) then alternative learning strategies such as deep neural models can be applied.  One challenge with such methods can be uncertainty quantification and the necessity of having very large data sets for model training.  Here I present an approach where we have integrated an alternative efficient “AI” spatio-temporal dynamical model, a deep echo state network, in a statistical framework to accommodate uncertainty quantification. This is applied to long-lead (one-year ahead) forecasting of settling patterns of Northern Pintails (Anas acuta) in the Prairie Pothole region of North America.

Related upcoming events

  • February 26, 2024 11:30 am - February 26, 2024 12:30 pm

    Out of Sight, Out of Mind

    Abstract: Join Bobby for a showcase of his detailed biological illustration, created under the pen name, NearsightGraphite. Highlighting his process, he’ll discuss the power of the observer and the illustrator’s role as a distiller of torrents of visual data. Bobby’s artwork amplifies intimate, obscure forms of life to give underrepresented organisms louder voices and to visualize their self-worlds. He aims to inspire curiosity in the intricacies of the natural world, and by extension, promote nature conservation.

    Bio: Robert J. Long ("Bobby") is a visual artist and educator from Silver Spring, Maryland. His zoologically-focused artwork has been exhibited throughout the United States and internationally. In 2013 he became Arizona’s first and only finalist in the internationally juried Adobe Design Achievement Awards. His illustrations have been published by Frontiers in Microbiology, Manifest Press, the Confluencenter for Creative Inquiry, and Edible Magazine, among others. Bobby is also an amateur reef aquarist and an avid photographer with experience in macro-photography and an interest in natural subjects.

  • March 4, 2024 11:30 am - March 4, 2024 12:30 pm

    Canopy Temperature

  • March 11, 2024 - March 15, 2024 All day
    No additional details for this event.
  • March 18, 2024 11:30 am - March 18, 2024 12:30 pm

    Carbon and Water Fluxes

  • March 25, 2024 11:30 am - March 25, 2024 12:30 pm

    Sustainability

  • April 1, 2024 11:30 am - April 1, 2024 12:30 pm

    Urban Ecology

  • April 8, 2024 11:30 am - April 8, 2024 12:30 pm

    Phenology

  • April 15, 2024 11:30 am - April 15, 2024 12:30 pm

    Kimberly Dibble, NPS

    KC Cushman, ORNL

    Soren Weber, ORNL

    Nicola Falco, Berkeley Lab

    Lucas Heintzman, USDA

    Cameron McIntire, USDA

     

  • April 22, 2024 11:30 am - April 22, 2024 12:30 pm

    Earth System Modeling

  • April 29, 2024 11:30 am - April 29, 2024 12:30 pm

    Microbes and Soil